This research is to investigate various statistical methods for analysis of censored survival data. In particular, examination of various nonparametric survival analysis techniques, their large sample properties, and related statistical tests and estimates are being studied. These include estimates for mean survival time, least squares regression estimates using censored data, and new statistical testing procedures are also under study. The methods are being examined for small sample properties by computer simulation and being applied to a large set of data on breast cancer patients.